4 ms·
It's a machine learning method for turning a sample of text into a fixed-length vector. We usually try to find embeddings with the property that samples of text
by soVeryTired 3y ago
It's a machine learning method for turning a sample of text into a fixed-length vector. We usually try to find embeddings with the property that samples of text with similar meanings are mapped to vectors that are close to one another.
- disqard 3y agoCould you recommend a self-hosted library for playing with text embeddings (as opposed to a REST API hosted online)? Thanks in advance!
- johnmoberg 3y agoYou could try SentenceTransformers: https://www.sbert.net/ https://www.sbert.net/
- disqard 3y agoThank you!
- kordlessagain 3y agoInstructor Large and XL work great.
- itronitron 3y agoWhy not just call it a text vector?